This study investigates the heart failure (HF) epidemic using community surveillance in a geographically defined population. In the current grant cycle (2003-06), we demonstrated that the incidence of HF, including in and out-patient cases, remained stable overtime while survival improved, thereby leading to an increasing prevalence of HF. This indicates that the HF epidemic is due to an increase in hospitalizations over time among a growing number of survivors. In this revised renewal application, we propose to investigate hospitalizations in HF and address the following questions: Is the increase in hospitalizations solely due to the increasing prevalence of HF in a growing population or is it related to an increasing number of recurrent hospitalizations per survivor? What are the risk factors for hospitalizations in HF, including determinants of recurrent events? This investigation should account for key methodological points, lacking from current literature: Temporal trends should be analyzed in an incidence cohort to avoid incidence-prevalence bias. Outpatient cases should be included as 42% of HF is diagnosed as outpatient. All-cause and HF-specific hospitalizations should be rigorously ascertained and examined in the same population. Indeed, HF-related hospitalizations relate to the effectiveness of the treatment of HF while all-cause hospitalization may reflect comorbidity among elderly patients with HF. Finally, analytical strategies should include the evaluation of multiple events. Two essential exposures should be prospectively captured: the type of HF, defined by ejection fraction (EF) and diastolic function given the heterogeneity of the HF syndrome (56% of prospectively ascertained patients in the community had normal EF and 64% of these have moderate or severe diastolic dysfunction) and psycho-social constructs, in particular health status. We propose to apply the rigorous surveillance methods implemented during the 1st grant cycle to address the following aims: Specific aim 1- temporal trends in all-cause and HF-specific hospitalizations after incident HF between 1979 and 2008. Specific aim 2- Association between type of HF (EF and diastolic function) and hospitalizations in a cohort of persons with validated HF, prospectively enrolled with an innovative approach using the electronic medical record that enables rapid ascertainment of cases of active HF. Specific aim 3- Association between health status and hospitalizations in HF prospectively measured using a generic instrument, the Short Form 12 and a disease specific instrument, the Kansas City Cardiomyopathy Questionnaire. The completion of these studies is uniquely possible in this setting because we will build on the proven methodology and findings of the initial funding period and capitalize on the infrastructure of the Rochester Epidemiology Project that captures in- and outpatient events, comorbid conditions and outcomes. Our team of investigators, strengthened by the addition of an expert in psychosocial constructs in HF, Dr. Moser, will employ novel and complementary design and analytical approaches that integrate passive surveillance and prospective identification of persons with HF at diagnosis in the same community. In doing so, we will gain important insights into the burden of hospitalizations among persons living with HF, which are a major driver of human, societal and heath care costs in HF. This will help define preventive strategies.